Learning from multiple frameworks for aquifer vulnerability mapping and multiple modelling practices in groundwater vulnerability mapping studies
Bibliographic record
Abstract
Learning from multiple frameworks (MF) in vulnerability mapping of aquifers and from multiple models (MM) is a novel research case tested in this paper by inclusive multiple modelling (IMM) practices. Each framework relates to multiple consensually selected data layers with an appropriate scoring system, which reflects intrinsic variances in the data layers and MF is particularly appropriate to shallow and patchy study areas. The IMM strategy is implemented at three levels: At Level 1, three frameworks (e.g., DRASTIC, SINTACS and GODS) are selected to map the vulnerability of a study area; At Level 2: inclusivity is achieved by employing the modelled output from Level 1 models as inputs for two additional machine learning models (e..g, support vector machine and multilayer perceptron) at Level 2. At Level 3: the outputs from these two models are combined using another model (e.g., random forest). The findings provide evidence that the Level 3 model produces more ‘defensible’ performance metrics by extracting information from all the models at Levels 1 and 2 with a better potential for learning from each output. The modelling results at Level 1 are ‘fit-for-purpose’, those at Level 3 are defensible and those at 2 are in between. For the patchy and shallow study area, the vulnerability maps at the higher level of the strategy are found to be more defensible than those at lower levels.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.026 | 0.051 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.005 | 0.008 |
| Open science | 0.003 | 0.008 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".